The long-term goal of this work is to develop strategies for using functional neuromuscular stimulation (FNS) of paralyzed muscles to enhance the recovery of individuals with incomplete spinal cord injury. The proposed work is motivated by three important developments. First, recent basic science and clinical studies have demonstrated that the degree of functional recovery of the injured spinal cord depends on the activity patterns of neural inputs to the spinal cord. Second, recent advances have produced adaptive controllers for FNS systems that provide a means of automatically adjusting stimulation parameters to reliably achieve specified rhythmic movements. Third, rodent models of spinal cord injury (complete and incomplete lesions) are extensively being used at the molecular, cellular, and systems level to investigate the effects of traumatic injury and to assess the results of therapeutic intervention. A combination therapy that utilizes locomotor training with FNS and pharmacological intervention is likely to be the most effective in enhancing the reorganization (plasticity) of the spinal circuitry that is spared after spinal trauma. A rodent model for FNS-assisted locomotion would facilitate quantitative evaluation of therapeutic regimens that include FNS and would provide the ability to characterize effects of FNS-assisted locomotion on the neuroanatomy and neurophysiology of the injured spinal cord. This biomedical engineering research grant proposal will develop a rodent model of locomotor training that utilizes treadmill walking and functional neuromuscular stimulation (FNS) with fixed-pattern and adaptive controllers. Kinematic and electromyogram (EMG) patterns of intact animals will be examined and then used to develop stimulation patterns for FNS-assisted movement. A series of tasks will be performed using FNS stimulation of hindlimb muscles in spinalized rats. These tasks will progress in difficulty from controlling suspended hindlimb movements to controlling hindlimb movements during treadmill locomotion in spinalized rats with partial weight support. Two different FNS control strategies will be used for each movement: a fixed-pattern, or open-loop, stimulation pattern and an adaptive stimulation control system. The adaptive stimulation control system will build upon our previous work and is expected to provide movement patterns that are more accurate and more repeatable. Successful completion of the proposed project will result in a novel animal model for FNS-assisted locomotor training and provide quantitative methods for evaluating locomotor behavior. In future studies, we plan to use a rodent model of incomplete spinal cord injury with FNS-assisted locomotion to test the hypothesis that FNS-assisted locomotor training enhances motor recovery after incomplete spinal cord injury. We anticipate that the improved performance provided by the adaptive control system may enhance the therapeutic effects of the technique. This locomotor training could also be combined with pharmacological intervention, tissue transplant, and neural repair therapies to determine if locomotor training can enhance the effectiveness of these therapies.